Llama 3.3

Latest

Meta

•

Open Source

•

70B

# 57

Released

Dec 6, 2024

# 13

Knowledge Cutoff

Dec 23

Benchmarks

# 90

Code RankedAGI

36.6%

# 43

Code LMArena

1232

# 51

Code LiveBench (old)

36.6%

# 66

GPQA Diamond

50.5%

# 30

Reason LiveBench (old)

50.7%

# 44

ELO LMArena

1257

# 37

Math LiveBench (old)

41.1%

# 20

MATH

77.0%

# 14

Human Eval

88.4%

# 2

Human Eval+

87.6%

# 22

NYT Connections

15.1%

# 38

MMLU Pro

68.9%

# 23

MMLU

86.0%

# 22

Halluc. Hughes

4.0%

# 18

Aidan Bench

779

# 5

IF LiveBench (old)

82.7%

# 27

Avg LiveBench (old)

56.7%

# 2

IF Evaluation

92.1%

Pricing

# 9

Input Cost /M

$0.1

# 11

Output Cost /M

$0.4